Google removes pledge not to use AI for weapons or surveillance from its ethical guidelines

The pledge introduced in 2018 not to use AI for applications 'likely to cause overall harm' has been removed.

Image via Google
2

In 2018, Google laid out a list of policies regarding its AI business that included safe and neutral use of the technology in its various applications. This week, Google has seemingly rolled back key policies regarding the use of its AI technology in weapons, surveillance, and other applications that would be “likely to cause overall harm.”

The change to Google’s AI principles was spotted via Washington Post, which pointed out the changes via a copy of Google’s AI Principles page that was preserved via the Internet Archive. An entire section of the principles, labeled “AI applications we will not pursue,” was removed from the redesigned webpage. This section included the above pledge not to use AI in a way that would cause harm, including for “weapons or other technologies whose principal purpose or implementation is to cause or directly facilitate injury to people.” Also removed was the use of “technologies that gather or use information for surveillance violating internationally accepted norms.”

Google's previous AI policies page including a section entitled
A previous version of Google's AI Principles page included a section called "AI applications we will not pursue," which had weapons and surveillance applications listed.
Source: Internet Archive

When asked about the changes, Google directed questions towards a recently written blog. Written by Google Research Labs SVP James Manyika and Google Deepmind CEO Demis Hassabis, the blog posits that with the expansion of AI technology in modern times, Google believes AI companies must be prepared to provide their tools to government and security clients.

It does little to say why Google itself would retract its own pledge to use its technology safely, but with the company continuing to pour billions into AI development, and seemingly benefiting heavily from it, the company has signaled that expansion is the priority. As we continue to follow AI, stay tuned for further updates.

Senior News Editor

TJ Denzer is a player and writer with a passion for games that has dominated a lifetime. He found his way to the Shacknews roster in late 2019 and has worked his way to Senior News Editor since. Between news coverage, he also aides notably in livestream projects like the indie game-focused Indie-licious, the Shacknews Stimulus Games, and the Shacknews Dump. You can reach him at tj.denzer@shacknews.com and also find him on BlueSky @JohnnyChugs.

From The Chatty
Hello, Meet Lola